Algorithm::Evolutionary::Experiment - Class for setting up an experiment with algorithms and population
use Algorithm::Evolutionary::Experiment; my $popSize = 20; my $indiType = 'BitString'; my $indiSize = 64; #Algorithm might be anything of type Op my $ex = new Algorithm::Evolutionary::Experiment $popSize, $indiType, $indiSize, $algorithm;
This class contains (as instance variables) an algorithm and a population, and applies one to the other. Can be serialized using XML, and can read an XML description of the experiment.
Creates a new experiment. An
Experiment has two parts: the population and the algorithm. The population is created from a set of parameters: popSize, indiType and indiSize, and an array of algorithms that will be applied sequentially. Alternatively, if only operators are passed as an argument, it is understood as an array of algorithms (including, probably, initialization of the population).
Creates a new experiment, same as before, but with an XML specification. An example of it follows:
<ea xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:noNamespaceSchemaLocation='ea-alpha.xsd' version='0.2'> <initial> <pop size='20'> <section name='indi'> <param name='type' value='BitString' /> <param name='length' value='64' /> </section> </pop> <op name='Easy' type='unary'> <param name='selrate' value='0.4' /> <param name='maxgen' value='100' /> <code type='eval' language='perl'> <src><![CDATA[ #source goes here ]]> </src> </code> <op name='GaussianMutation' type='unary' rate='1'> <param name='avg' value='0' /> <param name='stddev' value='0.1' /> </op> <op name='VectorCrossover' type='binary' rate='1'> <param name='numpoints' value='1' /> </op> </op> </initial> </ea>
This is an alternative constructor. Takes a well-formed string with the XML spec, which should have been done according to EvoSpec 0.3, or the same string processed with
XML::Parser::EasyTree, and returns a built experiment
Applies the different operators in the order that they appear; returns the population as a ref-to-array.
Opposite of fromXML; serializes the object in XML. First the operators, and then the population
Algorithm::Evolutionary::Run , another option for setting up experiments, which is the one you should rather use, as XML support is going to be deprecated (some day).
This file is released under the GPL. See the LICENSE file included in this distribution, or go to http://www.fsf.org/licenses/gpl.txt CVS Info: $Date: 2012/07/11 06:14:51 $ $Header: /media/Backup/Repos/opeal/opeal/Algorithm-Evolutionary/lib/Algorithm/Evolutionary/Experiment.pm,v 3.3 2012/07/11 06:14:51 jmerelo Exp $ $Author: jmerelo $ $Revision: 3.3 $